The infant index: a new outcome measure for pre-school children's services

The evaluation of community services for preschool children is hampered by the lack of valid and routinely available outcome measures. This study examines the use of data collected by teachers in response to educational legislation to determine whether a routine measure of attainments in primary school is sensitive to factors known to affect mental development.

A community child health dataset for the cohort of children born in Sheffield in 1990-1991 was matched with a dataset provided by schools in 1995-1996. The educational data consisted of the Infant Index scores which measure education attainments in reception class pupils.

We matched 4487 children from both datasets, which represented 75 per cent of all children born in the 1990-1991 cohort. Factors which predicted a poor Infant Index included male gender (odds ratio (OR) = 2.1, 95 per cent confidence interval (CI)= 1.8-2.6), low birthweight (OR = 1.4, 95 per cent CI = 1.1-1.9) and lack of breast feeding either by intention to feed (OR = 1.3, 95 per cent CI = 1.1-1.7) or actual feeding practice at one month (OR = 1.5, 95 per cent CI = 1.1-2.0). Other factors associated with a poor outcome for the child were postnatal depression, number of pregnancies, ethnicity, pre-school educational experiences and poor housing.

Although the results are interesting in themselves, the main significance of our project is in establishing a link between routinely collected health data and routine education data. This could facilitate research in the future thus leading to a considerable saving in the cost of long-term intervention studies.